基于接收信号强度的WSN优化定位算法研究

发布时间：2018-07-03 13:43

本文选题：接收信号强度 + 模型参数　；参考：《华中科技大学》2014年硕士论文

【摘要】：无线传感器网络中节点的部署环境通常比较复杂，导致节点间的通信模型无法预测，不同时刻节点间信号的传输方式存在很大的差异。正是传感器网络在实际环境中的这些动态性变化，以及季节，温度等对节点的影响，因此，需要提出一种对环境自适应求解模型参数的算法。仅仅通过环境中的实验数据或者经验模型来得到模型参数都是不准确的，本文旨在将当前环境下的实测数据特征和经验模型结合，设计参数自适应算法。同时，针对复杂环境中存在不可忽视的非视距误差，通过滤波器对接收信号强度值进行滤波。最后，基于参数自适应和非视距误差抑制模型设计一种适用于复杂环境的高精度定位算法。本文首先搭建了多组基于Mica2平台的实验，包括视距传播实验、单方向传播实验等，通过实验数据研究节点接收信号强度值的传播特性，同时分析对无线传播模型参数造成的影响。在对比现有传播模型参数求解的方法基础上，设计一种自适应求解模型参数的算法。将当前环境下的实测数据特征和经验模型相结合，保证模型参数更能适应环境的动态性变化。同时，针对复杂环境中非视距路径存在，导致节点间的测量值存在较大的非视距误差问题，比较现有非视距误差判定的方法，分析优缺点在此基础上进行改进，使之更适应于实际环境。并且设计有偏卡尔曼滤波器对非视距误差进行抑制，减小节点间的测距误差。最后，将粒子群优化算法与基于估计距离的加权质心算法结合，在对数-正态模型下进行定位效果仿真，，同时结合实际实验数据对算法进行定位性能分析，通过实验验证了该算法在复杂环境下得到了较好的定位效果。
[Abstract]:The deployment environment of nodes in wireless sensor networks is usually complex, which leads to the unpredictable communication model between nodes, and there are great differences in the transmission modes between nodes at different times. It is precisely the dynamic changes of sensor networks in the real environment and the effects of seasons and temperatures on the nodes. Therefore, an adaptive algorithm to solve the model parameters is proposed. It is not accurate to get the model parameters only from the experimental data or the empirical model in the environment. This paper aims to combine the characteristics of the measured data in the current environment with the empirical model to design an adaptive parameter algorithm. At the same time, the received signal strength is filtered by filter in view of the non-line-of-sight error which can not be ignored in complex environment. Finally, based on parametric adaptive and non-line-of-sight error suppression model, a high precision localization algorithm is designed for complex environment. In this paper, we first set up several experiments based on Mica2 platform, including line-of-sight propagation experiment, single-direction propagation experiment, etc. Through the experimental data, we studied the propagation characteristics of the received signal intensity value. At the same time, the influence on the parameters of wireless propagation model is analyzed. On the basis of comparing the existing methods to solve the parameters of the propagation model, an adaptive algorithm for solving the parameters of the model is designed. The characteristics of measured data and the empirical model are combined to ensure that the parameters of the model are more adaptable to the dynamic changes of the environment. At the same time, in view of the non-line-of-sight path existing in complex environment, which leads to the larger non-line-of-sight error problem between nodes, this paper compares the existing non-line-of-sight error determination methods, and analyzes the advantages and disadvantages of the improved method. Make it more suitable for the actual environment. A biased Kalman filter is designed to reduce the range error between nodes. Finally, the particle swarm optimization algorithm is combined with the weighted centroid algorithm based on the estimated distance, and the localization effect is simulated in logarithmic normal model. At the same time, the localization performance of the algorithm is analyzed based on the actual experimental data. The experimental results show that the algorithm has better localization effect in complex environment.
【学位授予单位】：华中科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN929.5;TP212.9

【参考文献】